With rapid development of computer and communication technology, a global information network is evolved to an Internet Protocol (IP)-based Next Generation Network (NGN). The amount of user equipment, service requirement and application scenario is increasing exponentially, therefore, the shortage of radio spectrum resource is further intensified. In this circumstance, Dynamic Network is proposed to explore reuse efficiency of radio spectrum resource by adaptively adjusting the network architecture, including functions of network nodes and data transmission paths and so on, during the network operation, and thus to satisfy the user requirement while guaranteeing effective resource utilization.
FIG. 1 is a schematic diagram of an example architecture of dynamic network. A wireless network generally consists of two kinds of logical entities, termed network infrastructure and user equipment (UE). As shown in FIG. 1(a), the network is formed by Base Stations (BSs) and UEs and a certain UE is only associated with some BS. During the network operation, the architecture of the network changes to be the one as shown in FIG. 1(b). The user equipment which provide signal relay service between base stations and other user equipment turn to be master user equipment (Master UE, mUE). The user equipment which associate with and obtain signal relay service from some mUE become slave user equipment (Slave UE, sUE). The user equipment which only associate with and obtain services from base station is called direct user equipment (Direct UE, dUE).
However, due to complexity and variability of the architecture, it makes network management difficult to involve dynamic network. An effective network management mechanism is required for dynamic network to decide whether a network architecture needs to be changed and how to change if required. But existing work provides neither performance evaluation of network architecture nor evaluation result based architecture design method for dynamic network.